Background: There is currently no validated questionnaire available to assess total sedentary time in older adults. Most studies only used TV viewing time as an indicator of sedentary time. The first aim of our study was to investigate the self-reported time spent by older persons on a set of sedentary activities, and to compare this with objective sedentary time measured by accelerometry. The second aim was to determine what set of self-reported sedentary activities should be used to validly rank people's total sedentary time. Finally we tested the reliability of our newly developed questionnaire using the best performing set of sedentary activities. Methods. The study sample included 83 men and women aged 65-92 y, a random sample of Longitudinal Aging Study Amsterdam participants, who completed a questionnaire including ten sedentary activities and wore an Actigraph GT3X accelerometer for 8 days. Spearman correlation coefficients were calculated to examine the association between self-reported time and objective sedentary time. The test-retest reliability was calculated using the intraclass correlation coefficient (ICC). Results: Mean total self-reported sedentary time was 10.4 (SD 3.5) h/d and was not significantly different from mean total objective sedentary time (10.2 (1.2) h/d, p = 0.63). Total self-reported sedentary time on an average day (sum of ten activities) correlated moderately (Spearman's r = 0.35, p < 0.01) with total objective sedentary time. The correlation improved when using the sum of six activities (r = 0.46, p < 0.01), and was much higher than when using TV watching only (r = 0.22, p = 0.05). The test-retest reliability of the sum of six sedentary activities was 0.71 (95% CI 0.57-0.81). Conclusions: A questionnaire including six sedentary activities was moderately associated with accelerometry-derived sedentary time and can be used to reliably rank sedentary time in older persons. © 2013 Visser and Koster; licensee BioMed Central Ltd.